Greek, Roman, and Byzantine Studies 45 (2005) 131–153 © 2005 GRBS The Eastern Mediterranean Epic Tradition from Bilgames and Akka to the Song of Release to Homer’s Iliad Mary R. Bachvarova N 1983 the first pieces were found of a new story, the bi- lingual Hurro-Hittite Song of Release, in Houses 15 and 16 of the “Oberstadt” of Hattusa (Boğazköy). Martin West was unable to take much of the text into account in his masterly study of the connections between Near Eastern and Greek poetry, The East Face of Helicon, because the complete edition of the work by Erich Neu appeared only in 1996.1 These tablets 1 M. L. West, East Face of Helicon (Oxford 1997) 75 n.60, did mention the parables in the story. The text is edited by E. Neu, Das hurritische Epos der Freilassung I: Untersuchen zu einem hurritisch-hethitischen Textensemble aus Hattuš̌a (Wiesbaden 1996), who provided a translation of all the coherent fragments. All citations from the Song of Release follow Neu’s edition unless otherwise noted. All translations are my own. A near complete translation into English may be found in H. A. Hoffner, Jr., Hittite Myths2 (Atlanta 1998) 65–80; partial translations are G. Wilhelm, “The Hurrians in the Western Parts of the Ancient Near East,” Michmanim 9 (1996) 17–30, G. Beckman in W. W. Hallo and K. L. Younger (eds.), The Context of Scripture I Canonical Compositions from the Biblical World (Leiden 1997) 216–217, A. Ünal in K. Hecker et al. (eds.), Mythen und Epen II (Texte aus der Umwelt des Alten Testaments 3.4 [Gütersloh 1994]) 860–865, and G. Wilhelm in M. Dietrich et al. (eds.), Ergänzungslieferung (Texte aus der Umwelt des Alten Testaments [Gütersloh 2001]) 82–91. My own interpretation differs in some important details however from these works. I defend this interpretation in “Relations between God and Man in the Hurro-Hittite ‘Song of Release,’” JAOS (forthcoming), using phraseological and thematic correspondences between the Song of Release and various Hittite texts (treaties, administrative texts, annals, prayers, and other Hurro-Hittite songs) to argue that the Song of Release is not about releasing debt slaves (so Hoffner and others) but about releasing captives. This point has already been argued from a different angle by E. Otto, “Kirenzi und derôr in der hurritisch-hethitischen Serie ‘Freilassung’ (parā � I 132 THE EASTERN MEDITERRANEAN EPIC TRADITION will in fact prove very valuable to all those interested in study- ing the development of ancient “epic”—traditional narrative poetry about gods and men—around the Mediterranean during the Late Bronze Age. Even the few fragments that we have are filled with “Homeric” motifs drawn from the same Near Eastern literary tradition which apparently was familiar to many Greek poets in some form or another; the Song of Release can thus shed new light on how the themes, plot lines, motifs, and formulae of courtly epic could have been translated and transmitted from the larger Near East to Greek-speakers, providing us with an example of how the wider Near Eastern epic tradition was transported across linguistic barriers and adapted to the particular interests of a new milieu. The tablets of the Song of Release are laid out with the Hurrian original on the left side and the Hittite translation on the right. Among the fragments a few colophons are preserved which tell us that the text was a SIR3 par ā tarnumaš̌ “Song of Release” (Keilschrifttexte aus Boghazköi [KBo] 32.11, 13, 15, 19, 66), per- formed (at some point in its history) by a LÚNAR, a male singer (colophons of 13, 66).2 The SIR3, or “song,” was a Hittite genre that included songs about Kumarbi, Silver, Gilgamesh, and Keš̌š̌i, and the Song of Release uses phraseology and motifs found in these works.3 The epic motifs discernable in the Song of ___ tarnumar),” in G. Wilhelm (ed.), Akten des IV. Internationalen Kongresses für Hethitologie (Wiesbaden 2001) 524–531. 2 See E. Neu, “Knechtschaft und Freiheit,” in B. Janowski, K. Koch, and G. Wilhelm (eds.), Religionsgeschichtliche Beziehungen zwischen Kleinasien, Nord- syrien und dem Alten Testament (Göttingen 1993) 329–361, at 330–331, and M. Schuol, “Zur Überlieferung homerischer Epen vor dem Hintergrund alt- anatolischer Traditionen,” in M. Schuol, U. Hartmann, and A. Luther (eds.), Grenzüberschreitungen: Formen des Kontakts zwischen Orient und Okzident im Altertum (Stuttgart 2002) 331–363, at 336–340. Some five or six tablets in total present coherent passages. Fifteen or sixteen more smaller pieces give us tantalizing glimpses of other parts of the narrative and about 100 more even smaller pieces add little else to our knowledge, but indicate that the text was very repetitious. 3 M. Bachvarova, From Hittite to Homer: The Role of Anatolians in the Trans- mission of Epic and Prayer Motifs from the Ancient Near East to the Ancient Greeks (diss. Univ. Chicago 2002) 120–128, and E. Neu, “Keš̌ š̌e-Epos und Epos der Freilassung,” SMEA 31 (1993) 111–120. MARY R. BACHVAROVA 133 Release include the proemium (11), the assembly scene (15, 16, 19), a hospitality sequence (13, 46, 65, 72, 209),4 and waking for a message (37).5 This paper focuses primarily on the assembly scene in the light of other Near Eastern narratives and the Iliad, comparing in detail the assembly scene of the Song of Release with the Iliad and then briefly discussing parallels from the Sumerian Bilgames and Akka, 1 Kings 12 from the Bible, and the Old Babylonian creation epic Atrahasis. This comparative discussion shows how motifs were reworked and inverted to create new plots from traditional material and demonstrates that the assembly scene in the Song of Release stands midway between Greek and Near Eastern epic, for the very Greek motif of two human speakers arguing before an assembly of other humans, although built up out of stock Near Eastern motifs, does not appear with all its parts in any other Near Eastern narrative poem. One tablet of the Song of Release is nearly complete; it contains a series of parables in the tradition of Aesop’s fables and A Thousand and One Nights, a tradition with biblical parallels as well.6 The narrative as a whole seems to be divided between parables and a less well preserved plot line telling of the fall of Ebla. As yet it is unclear how these parables are connected to the rest of the plot, and Wilhelm has even suggested that they do not belong to the same text at all.7 The destruction of the town seems to be blamed on the divine wrath of the Hurrian storm-god Teš̌š̌ub, angered when the Eblaites refused to release certain captives, people of the town Ikinkalis. This town in North Syria was conquered in approximately 1600 B.C. by the Old Hittite king Hattusili I or his successor Mursili I.8 The song 4 See Bachvarova, Hittite to Homer 107–116; S. Reece, The Stranger’s Wel- come: Oral Theory and the Aesthetics of the Homeric Hospitality Scene (Ann Arbor 1993); West, East Face 201–203; and compare especially the embassy to Achilles in Iliad 9.185–224. 5 Bachvarova, Hittite to Homer 116–120; West, East Face 185–193, 229; I. McNeill, “The Metre of the Hittite Epic,” AnatSt 13 (1963) 237–242, at 241. 6 See Hoffner, Hittite Myths 70. 7 In Dietrich, Ergänzungslieferung 84–85. 8 See H. Klengel, Syria: 3000 to 300 B.C.: A Handbook of Political History (Berlin 1992) 80–83. 134 THE EASTERN MEDITERRANEAN EPIC TRADITION thus may be interpreted as providing an after-the-fact justifica- tion for the conquest.9 As Neu himself pointed out, the Hurrian version of the song could have been imported during the Old Hittite period, when the Hittites were actively campaigning in North Syria and conquered both Ebla and Ikinkalis, although its palaeography and grammar indicate that the song was writ- ten down and translated into Hittite in the Middle Hittite period (1500–1400 B.C.).10 As has been pointed out already, the poem opens in a man- ner typical of Eastern Mediterranean hymns and epics from the Sumerian Ur III Hymn to Gilgamesh to the Akkadian Creation Epic to the Hurro-Hittite Song of Silver to the Homeric hymns and epics (KBo 32.11 i 1–13, preserved only in Hurrian):11 I shall sing of Teš̌š̌ub, great lord of Kummi I shall exalt the lady Allani, at the doorbolt of the earth And along with them I shall tell of the lady Ishara, the word-maker, speaker of wisdom, god. 9 See V. Haas and I. Wegner, “Baugrube und Fundament,” IstMitt 43 (1993) 57, and E. Neu, “La bilingue hourro-hittite de Hattusha, contenu et sens,” in J.-M. Durand (ed.), Amurru: I Mari, Ebla et les hourrites: Dix ans de travaux (Paris 1996) 189–195, at 192–193. 10 Neu, Epos 483 and 3–7. 11 H. A. Hoffner, Jr., “The Song of Silver: A Member of the Kumarbi Cycle of ‘Songs,’” in E. Neu and C. Rüster (eds.), Documentum Asiae Minoris Antiquae: Festschrift für Heinrich Otten (Wiesbaden 1988) 143–166; S. de Mar- tino, “Il ‘Canto della liberazione’: Composizione litteraria bilingue hurrico- ittita sulla distruzione de Ebla,” ParPass 55 (2000) 296–320, at 300–301; and Schuol, Grenzüberschreitungen 331, 336–337. Also see C. Wilcke, “Die An- fänge der akkadishen Epen,” ZA 67 (1977) 153–207, at 153–155, 175–186, 200–202; West, East Face 170–173; and B. De Vries, The Style of Hittite Epic and Mythology (diss. Brandeis 1967) 127–134, on proemia. A new fragment of the opening of the Standard Babylonian Gilgamesh epic shows that it does not contain the “I sing …” opening which has frequently been restored for it: T. Kwasman, “A New Join to the Epic of Gilgameš Tablet I,” Nouvelles assyriologiques brèves et utilitaires 89 (1998) #99, and transl. of A. George, The Epic of Gilgamesh: The Babylonian Epic Poem and Other Texts in Akkadian and Sumerian (London 1999) 1. MARY R. BACHVAROVA 135 I shall tell of Pizikarra … Ebla … who will bring … Pizikarra will des[troy (?) … Nuha š še … and Ebla … Pizikarra, the Ninevan … he bound (participle, absolutive case) … bound … … with the gods … Between the opening preserved in KBo 32.11 and the assembly scene a series of parables told by an intradiegetic narrator seems to intervene. These warn of the consequences of in- subordination, greed, and arrogance, but the advice proffered does not seem to have been followed by the characters in the main narrative. Perhaps they may be compared to the story told by Phoenix and ignored by Achilles in the embassy to Achilles in Iliad 9.531–605.12 The key scene to be discussed here appears in KBo 32.16 and 15. A short passage in 16 introduces us to one of the pro- tagonists, Zazalla, the opponent in assembly of the king Meki. The rest of 16 is very damaged, but the beginning of 15, the fifth tablet of the Song of Release, duplicates the end of 16. Here Zazalla seems to be the one who argues that if Teš̌š̌ub were in want, each would contribute whatever he might need; if someone were depriving the god, the Eblaites would mitigate Teš̌š̌ub’s suffering (KBo 32.15 ii 4'–9'): [If Teš̌] š̌ub is injured by oppression and he [a]sks [for release], if Teš̌š̌ub [is o]ppressed, each will g[i]ve to Teš̌š̌ub [one shekel of silver.] Ea[ch] will give half a shekel [of gold], [we will each g]ive to h[im] of silver [one shekel]. But, if he, Teš̌š̌ub, is hungry, we will each give one measure of barley [to the g]od. 12 For an interpretation of the text that unites the parable section and the Ebla section under the single theme of insubordination, see Otto, Akten 524– 531. 136 THE EASTERN MEDITERRANEAN EPIC TRADITION Zazalla speaks harshly to Meki (ii 20'–24'): For you, Meki, does your heart rejoice inside? First of all, for you, Meki, your heart inside will not rejoice. Secondly, for Purra, who is to be given back, his heart inside will rejoice. Under no circumstances however will the Eblaites let Purra and the sons of the town Ikinkalis go, because they and their kings need them to do their menial labor. Meki should send his own slaves and his own wife and children (ii 26'–iii 6): Why will we let them go And who will give us food? They are our cupbearers, and they give out. They are our cooks, and they wash for us. And the thread which they spin is [thick] like the hair [of an ox.] But if for you releasing [is desirable,] re[lease] your male and female servants! Surrender your son! [Your] wife […] send! (…) Meki seems helpless against Zazalla. He turns to Teš̌š̌ub, falling at his feet and telling of his city’s refusal and trying to avoid any blame—and here our tablet breaks off (iii 13–20): “Listen to me Teš̌š̌ub, great king of Kummi. I will [gi]ve it, (i.e.) pariš̌š̌an, but m[y c]ity will not give it. And Zazalla, son of Pazz[anik]arri, won’t give release.” Meki (tried to?) purify his ci[ty] from sin, the ci[ty of Eb]la. He (tried to?) waive the sins for the sake of his city … MARY R. BACHVAROVA 137 I translate the preterites parkunut (19) and peš̌š̌iyet (20) as “(tried to?)” because it seems that Meki is unable to prevent the de- struction of his city when the Eblaite assembly refuses to release the captives, as Teš̌š̌ub himself had demanded (KBo 32.19–20). The introductory descripton of Zazalla presents interesting parallels to the stock characterization of speakers in the Iliad (KBo 32.16 ii 1–13 with 32.54, 59): [… there is no] one who speaks against him [ …] among the elders, there is no one who speaks against him. [There is no one] who makes a response to him. [No] one talks. But if there is one who speaks greatly in the city, whose words no one turns aside, Zazalla is the one who speaks greatly. In the place of assembly, his words no one overcomes. [Zazal]la began to speak to Meki, “Why [do you] speak humility, star of Ebla, Me[ki]…?” This description of Zazalla is similar to that of Thoas in the Iliad (15.283–284): “In the assembly few of the Achaean bested him when the young men competed with words” (égorª d° • paËroi ÉAxai«n n¤kvn, ıppÒte koËroi §r¤sseian per‹ mÊyvn). The comment of Nestor to Diomedes is equally apropos (9.54– 56):13 ka‹ boulª metå pãntaw ımÆlikaw ¶pleu êristow. oÎ t¤w toi tÚn mËyon ÙnÒssetai ˜ssoi ÉAxaio¤, oÈd¢ pãlin §r°ei in council among all your peers you are the best, no one will fault your speech, as many are the Achaeans, nor will (any one) speak against it. 13 On this formulaic description of the able speaker see K. Dickson, Nes- tor: Poetic Memory in Greek Epic (New York 1995) 12. All quotations from the Iliad follow Monro and Allen’s text (Oxford 1920). 138 THE EASTERN MEDITERRANEAN EPIC TRADITION Zazalla is in fact the first human example of the “excellent speaker in assembly” that I have found. But whereas Nestor and the other great Greek speakers were conciliators speaking the truth, Zazalla is antagonistic, arguing against king Meki and in the wrong. In some ways he is more like the famed Iliadic fluent but unwise speaker Thersites, who attempts to turn the assembly of Argives against their leaders, criticizing Agamemnon.14 Like Zazalla, Thersites dares to speak rudely to a leader of the people. But the positions of the two speakers are inverted in relation to each other. In contrast to Zazalla, Thersites speaks the truth but too plainly, com- plaining that Agamemnon should be satisfied with what he has already won and that he has gone too far in insulting Achilles (Il. 2.225–242). Thersites fails to persuade the assembly and is silenced ignominiously by Odysseus. Zazalla, on the other hand, is respected by the assembly of Ebla. He is able to sway the assembly to his side despite Meki’s opposition, yet his ad- vice proves to be wrong. We turn now to the beginning of the Iliad to see how the themes that appear in the debate scene of the Song of Release are reworked for a Greek audience. The Iliad opens with two consecutive debates in an assembly over freeing Chryseis, the daughter of a local priest of Apollo. The priest first offers a ransom in return, but Agamemnon refuses his request. In this narrative sequence appear many of the motifs found in the assembly scene of the Song of Release: a request to release a captive which is refused, a pleader in contact with his god, an assembly, and an unwise speaker who carries the day to his people’s detriment. In the Song of Release, a man in close contact with his god, the king Meki, seems to present the god’s request in assembly that captives should be freed, but the assembly is persuaded by another of its leaders to go against the word of the king and keep the captives to work for them. In the Iliad, on the other hand, the captive is the pleader’s own daughter, and the pleader is the one who will rouse the god to anger on his behalf. When the priest of Apollo comes to plead his case, he does not serve as the god’s representative. Only after the priest 14 See Il. 2.212–224 and discussion in Dickson, Nestor 27, 51–52. MARY R. BACHVAROVA 139 is rejected does Apollo become involved at his request. It is Chalcas at the instigation of Achilles who makes an attempt to divine the god’s thoughts and states correctly that he is angered specifically over the captive Chryseis. The Greek assembly ex- presses its opinion, but has no control over the final decision. Meki, on the other hand, wishes to avoid the wrath of Teš̌š̌ub, who himself demands the release of the captives. King Aga- memnon, going against the opinion of his assembly, refuses the priest’s request, saying that he enjoys the captive woman’s ser- vices, and sends the priest away harshly (1.26–32): mÆ se, g°ron, ko¤l˙sin §g∆ parå nhus‹ kixe¤v µ nËn dhyÊnont' µ Ïsteron aÔtiw fiÒnta, mÆ nÊ toi oÈ xra¤sm˙ sk∞ptron ka‹ st°mma yeo›o: tØn d' §g∆ oÈ lÊsv: pr¤n min ka‹ g∞raw ¶peisin ≤met°rƒ §n‹ o‡kƒ, §n ÖArgeÛ, thlÒyi pãtrhw, flstÚn §poixom°nhn ka‹ §mÚn l°xow éntiÒvsan: éll' ‡yi, mÆ m' §r°yize, sa≈terow Àw ke n°hai. Let me not meet you, old man, by the hollow ships, either lingering now or coming back again in the future, lest your staff and fillet of god fail to protect you. I will not release her, rather old age will come upon her in my house in Argos, far from her fatherland, going back and forth before the loom, and sharing my bed; but go, don’t annoy me, so that you may go safer. In the Iliad Chryses then calls upon Apollo, using the same persuasive techniques that his Near Eastern colleagues would, reminding him of the services he has provided the god.15 The plague sent by Apollo devastates the Greek host, and finally another assembly is called to ferret out the source of impurity that has provoked this disaster at the hands of Apollo. Now Achilles attempts to address the situation in terms very similar to those a Hittite or an Akkadian would use, wondering at first if Apollo feels slighted by the Greeks’ neglect, and re- 15 Il. 1.33–42, and see West, East Face 348, 273. 140 THE EASTERN MEDITERRANEAN EPIC TRADITION questing that an omen be sought (Il. 1.53–67).16 It is revealed that Apollo is angered over the insult to his priest, and Achilles argues successfully against Agamemnon, forcing him to give the girl up. Agamemnon however insists that he be compen- sated for the loss of his concubine, and threatens to take one from someone else (131–139): mØ d' oÏtvw, égayÒw per §∆n, yeoe¤kel' ÉAxilleË, kl°pte nÒƒ, §pe‹ oÈ pareleÊseai oÈd° me pe¤seiw. ∑ §y°leiw ˆfr' aÈtÚw ¶x˙w g°raw, aÈtår ¶m' aÎtvw ∏syai deuÒmenon, k°leai d° me t∞nd' épodoËnai; éll' efi m¢n d≈sousi g°raw megãyumoi ÉAxaio¤, êrsantew katå yumÒn, ˜pvw éntãjion ¶stai: efi d° ke mØ d≈vsin, §g∆ d° ken aÈtÚw ßlvmai µ teÚn µ A‡antow fi∆n g°raw, µ ÉOdus∞ow êjv •l≈n: ı d° ken kexol≈setai ˜n ken ·kvmai. Don’t in this way, noble though you may be, god-like Achilles, try to deceive with cleverness, since you will neither divert me nor persuade me. Do you indeed wish, while you yourself hold a prize, for me to sit here lacking one? Do you order me to give her back? But if the great-hearted Achaeans will give the prize, choosing according to wish, so that it will be compensation— if they don’t give, I myself will go choose either your prize or, Ajax’, or Odysseus’, I will take and lead away; he will be angered, whom I visit. The subject and phrasing of Agamemnon’s assertions to Chryses and Achilles have interesting similarities to Zazalla’s declaration in the Song of Release. Agamemnon declares to Achilles that he will take another’s captive in compensation, just as Zazalla suggests angrily that Meki give up his own ser- 16 Compare the Hittite Kantuzzili’s Prayer, transl. I. Singer, Hittite Prayers (Atlanta 2002) 31–36, itself based on Mesopotamian hymns to the Sun-god, and the Akkadian Poem of the Righteous Sufferer, transl. B. Foster, Before the Muses: An Anthology of Akkadian Literature (Bethesda 1993) 308ff., and see the discussion in West, East Face 124–128. MARY R. BACHVAROVA 141 vants, even his wife and child. Just as Zazalla describes the chores of the people of Ikinkalis, Agamemnon describes the tasks that Chryseis fulfills for him. Just as Agamemnon speaks harshly to the priest (1.32) and Achilles (1.139), Zazalla speaks rudely and sarcastically to Meki and with him to the chief cap- tive, Purra. Perhaps it is only natural that a dispute between two power- ful antagonists would play itself out with such verbiage, but the premise of the Iliadic dispute is similar in many details to the situation in the Song of Release, as far as we can make out from the latter’s frustratingly fragmentary remains.17 However, the reasons for the detention and release of the person in question are culture-specific. In the case of the Greeks the reason is clear. Chryseis is a war-captive selected as a prize by Aga- memnon. Obtaining women by capture, whether for marriage or to be a concubine, was part of Greek culture. When Herodotus (1.1–5) discusses the beginning of Greek-barbarian relations, for example, he sees it as a series of retaliatory con- flicts over stolen brides including Helen. The ultimate cause for the Trojan War is always in the background of the conflict that opens the Iliad, which simply replays on a smaller scale the reason for the larger conflict. Hittites sometimes did obtain brides by capture also, as evinced by Laws §37.18 However, there is no hint of that in what we have of the Song of Release. The people of Ikinkalis are certainly not stolen brides. Although the suggestion of other scholars that the captives from Ikinkalis were debt slaves has now been debunked by Otto, the song does seem to be adapt- 17 Compare the threatening words of Anu in the Hurro-Hittite Song of Kumarbi (§6, transl. Hoffner, Hittite Myths 43), also Gilgamesh (Standard Bab- ylonian Version V 100–101, transl. George, Epic 41), the Sumerian Enmerkar and the Lord of Aratta (226–227, transl. T. Jacobsen, The Harps That Once…: Sumerian Poetry in Translation [New Haven 1987] 295), and the Ugaritic story of Baal (Die Keilalphabetischen Texte aus Ugarit 1.2 i 24–26, transl. N. Wyatt, Religious Texts from Ugarit: The Words of Ilimilku and his Colleagues [Sheffield 1998] 60). Arguments and verbal conflicts in Homer are studied by Dick- son, Nestor 111–117. 18 See discussion in H. A. Hoffner, Jr., The Laws of the Hittite: A Critical Edi- tion (Leiden 1997) 187. 142 THE EASTERN MEDITERRANEAN EPIC TRADITION ing a Near Eastern motif of a divine demand for the release of debt slaves.19 The most likely possibility is that the servants were taken captive in a previous altercation between the men of Ikinkalis and Ebla, or that Ikinkalis as a whole was subor- dinate to Ebla. In the Song of Release, it is unclear how the freeing of the people of Ikinkalis is connected to assuaging the physical needs of Teš̌š̌ub through sacrifices, or indeed whether Teš̌š̌ub is in fact suffering and in need of offerings, or the description of the miserable state of the god is merely hypothetical. However, Zazalla says that the primary reason he will not release the cap- tives is that he would then have to do himself the menial labor in which they have been employed, i.e., work to fulfill his own physical needs for food and clothing, and Teš̌š̌ub himself demands the release of the servants. Further, other Hittite doc- uments show that sometimes people, whether war captives or free Hittites, were released from compulsory labor in order to devote them to serving a god.20 Finally, in the Hurro-Hittite Song of Hedammu, part of the Kumarbi cycle, the god Ea is con- cerned that if the gods allow the destruction of human worshippers, they will be forced to do menial labor (§6).21 It is possible to see Meki’s attempts to purify his city from sin and cast it away as an attempt at compensation paid to the god, just as Zazalla’s hypothetical offers of payment to Teš̌š̌ub could be seen as compensation. However, it does not seem to be an all- pervasive theme in the Song of Release. In the Iliad, however, compensation is a recurrent theme, relating in the end to the idea of glory in the form of epic as compensation for the early death of the Homeric warrior im- 19 Neu, Epos 9, and Hoffner, Hittite Myths 180–183, argue that the people of Ikinkalis are debt slaves, an interpretation corrected by Otto, Akten 524– 531. Bachvarova, JAOS (forthcoming), argues that the plot is an adaptation of a Near Eastern tradition. 20 Annals of Hattusili I, KBo 10.2 iii 15–20, ed. and transl. F. Imparati and C. Saporetti, “L’autobiografia di Hattuš̌ili I,” Studi classici e orientali 14 (1965) 40–85, at 52–53; decree of queen Asmunikal, transl. H. Otten, Hethitische Totenrituale (Berlin 1958) 106; Apology of Hattusili III §13, transl. T. van den Hout in Hallo/Younger, Context 204. 21 Transl. Hoffner, Hittite Myths 52. MARY R. BACHVAROVA 143 mortalized in it.22 The court scene on Achilles’ shield (18.497– 508) shows two speakers debating the possibility of a murderer atoning for his crime with a monetary payment rather than his life, thus showcasing debates over compensation as one of the essential public performances for Greeks. The action of the Greek epic is set into motion by two demands for compensa- tion for lost captives, Chryseis and Briseis, and this debate among the Greeks is matched with a debate among the Trojans over returning Helen in 7.345–378, and the earlier meeting that had occurred outside of the action of the Iliad, mentioned at 3.204–224 and 11.122–142, in which members of the council of elders, because they had been bribed by Paris, refused to return Helen to Menelaus and Odysseus.23 When the Achaeans reach the conclusion that Apollo is indeed angered over the mistreatment of his priest, compensation is in fact paid, but to Chryses not Apollo, although the Achaeans do make sure to propitiate Apollo as well when Chryseis is re- turned (1.446–474). Thus, although the Iliad is reusing a traditional narrative sequence of a dispute in assembly over freeing people from forced labor, it is reworking the components to address the paramount concern of the Homeric warrior, his honor, and to echo the larger narrative sequence of the capture (or escape) and return of Helen. In the Iliad the return of Chryseis is demanded by her father, motivated by filial affection. Thus, the Near Eastern theme of freeing temple personnel or debt slaves in order to serve the god in question is realized rather differ- ently in Greek, as freeing the daughter of the priest of a god who intervenes on his behalf. Yet the idea of the reciprocal re- lationship between man and god found in the Near Eastern material is continued, with the priest Chryses pointing out to 22 See G. Nagy, The Best of the Achaeans: Concepts of the Hero in Archaic Greek Poetry2 (Baltimore 1999) 28–29, on the idea of glorification in epic as the rec- ompense for dying young. Near Eastern parallels are discussed in West, East Face 373. The theme of compensation, which links thematically the stories of Achilles and Helen, has been discussed by D. F. Wilson, Ransom, Revenge, and Heroic Identity in the Iliad (Cambridge 2002). 23 This is discussed by W. M. Sale, “The Government of the Trojans,” GRBS 30 (1989) 1–100, at 65–80. 144 THE EASTERN MEDITERRANEAN EPIC TRADITION Apollo that he has faithfully fed and housed him; this theme is combined with the Homeric theme of obtaining women by capture. The offence originally committed when Agamemnon refuses to release Chryseis is to the honor of Chryses, who cannot retaliate because he is just a weak old man. Furthermore, Agamemnon slights Apollo by insulting his priest; when Apollo enters the scene he is supporting a loyal servant. Agamemnon sees Achilles’ insistence on returning the girl to her father as an assault on his own honor and retaliates by injuring him in turn. The over-arching plot line of the return by force of Helen is motivated not by the thought that she is unjustly imprisoned but by the need to repair the honor of Menelaus. The Song of Release puts in a new light the question of the historicity of the Iliad. Scholars have accepted that there could have been one or more conflicts between Greek-speakers and Trojans, but have been more skeptical of the possibility of a war being motivated over the kidnapping of a queen. As Trevor Bryce puts it:24 There are those who firmly maintain, perhaps not without justification, that the war was fought over the abduction of a Mycenaean queen, even if she were a willing abductee. Hittite kings were certainly prepared to go to war to reclaim subjects who had been removed, whether forcibly or voluntarily, from their kingdom. But all speculations about the possible reasons for a Greek-Trojan conflict bring us back to the basic question of whether the tradition of the Trojan War has an authentic historical basis. The Song of Release shows us that the historical kernel of fact be- hind the Iliadic conflict, if there was one, was immortalized in a traditional story line concerning a conflict over captives that could explain the destruction of a city. In the Song of Release, the consequences of Zazalla’s arrogance provide us with the message, “Do not go against the word of the king.” While there is some indication that the Hittite king 24 The Kingdom of the Hittites (Oxford 1998) 398. MARY R. BACHVAROVA 145 did discipline his assembly in regard to their conduct towards inferiors,25 we are not yet sufficiently informed with respect to the real-life context that gave meaning to the conflict in the Hurro-Hittite epic to fully understand how its original audience would have perceived it. But we can at least see that the story supports the king of Ebla against Meki and the Eblaite as- sembly. In the Iliad Agamemnon is pitted against Achilles, each rep- resenting two opposing types of leader. Both are kings by birth, but Achilles is the better warrior while Agamemnon leads the largest contingent. The Iliad is ambivalent with regard to the legitimacy of Agamemnon’s rule. On the one hand, Thersites presents valid arguments against Agamemnon, yet he is beaten and silenced by Odysseus to the approval of the rest of the as- sembly. Meanwhile, Agamemnon is portrayed so negatively that Dean Hammer has suggested that the Iliad is “a reflection on the nature of political authority,” caused by:26 a fundamental shift in the type of political questions asked, from the “power of authority” to carry out decisions suggestive of Dark Age politics to the legitimacy of authority in making these decisions, a question critical to the formation of an increasingly interdependent polis form of political organization. Despite the fact that Thersites is silenced, he seems to represent a legitimate voice of dissent against the aristocracy, one ex- pressed by a variety of archaic Greek poets.27 There are two good parallels in the wider Near Eastern lit- erature to the assembly scene in the Song of Release, one in 1 Kings 12:1–20 and one in the Sumerian epic Bilgames and Akka. Yet in neither of these examples is the king rebuffed and overruled as Meki was by his assembly; rather the king is free to make his own decision. These examples make clear how 25 See KBo 22.1 16–20, transl. G. Beckman, “The Hittite Assembly,” JAOS 102 (1982) 435–442, at 440–441. 26 D. Hammer, “‘Who Shall Readily Obey?’: Authority and Politics in the Iliad,” Phoenix 51 (1997) 1–23, at 2. 27 See W. Donlan, “The Tradition of Anti-Aristocratic Thought in Early Greek Poetry,” Historia 22 (1973) 145–154. 146 THE EASTERN MEDITERRANEAN EPIC TRADITION striking this plot twist in the Song of Release must have been to the Hittite audience. Gilgamesh argues directly with his as- sembly of elders over the imposition of corvée labor on his people by king Akka of Kish. He succeeds in winning over the young men of Uruk in a second assembly, who then go to fight against Akka (1–47).28 At 1 Kings 12:1–20 after King Solomon dies, his son Rehoboam is about to become king, and Jero- boam, who had been promised that he would some day be the leader of most of the tribes of Israel, returns from Egypt and asks Rehoboam in concert with the people of Israel who have gathered in assembly if Rehoboam will be more merciful towards them than his father was. While the older men en- courage Rehoboam to be kind, the younger men advise that he reply harshly. Rehoboam follows the bad advice of the younger men, since God wished it so in order to fulfill his promise to punish Solomon for no longer worshipping him properly, and to honor Jeroboam; as a result, Israel rebels against Reho- boam. In the biblical example we have two sides of a discussion presented in an assembly over whether to be merciful to an oppressed and overworked people. The “wrong” argument is couched in angry and sarcastic language similar to that of Agamemnon and Zazalla. The wrong decision is made, to con- tinue to oppress them, and that leads to the downfall of the leader who made the decision. However, unlike Meki, Reho- boam does not argue for a particular side, and is in control of the final decision. A further Near Eastern text that shows the close parallels to the Song of Release is the Akkadian Atrahasis, although it describes an assembly of gods, not humans. Atrahasis tells the story of the 28 See transl. in George, Epic 145–147. Further discussion of the role of the assembly in this poem may be found in D. Katz, “Gilgamesh and Akka: Was Uruk Ruled by Two Assemblies?” RAssyr 81 (1987) 105–114, and R. T. Riley, “The Saga of an Epic: Gilgamesh and the Constitution of Uruk,” Orientalia 69 (2000) 341–367. D. E. Fleming, Democracy’s Ancient Ancestors: Mari and Early Collective Government (Cambridge 2004), is the most recent dis- cussion of real-life assemblies in the Near East; see esp. 204–206, where he discusses the literary passages covered in this section. Also compare 2 Samuel 16:15–17:14, the dispute of Hushai and Ahithophelum before “all the men of Israel”; I owe this reference to Prof. H. A. Hoffner, Jr. MARY R. BACHVAROVA 147 invention of mankind to do manual labor for the gods and the gods’ attempt to destroy them, when they become too numer- ous, with a flood which destroys all of mankind except for the clever Atrahasis and those he brought on board; they are saved because he is careful to follow the advice of the god Enki. The flood story is similar to the Old Testament Noah story or the Greek story of Deucalion.29 The Akkadian narrative is a product of a long-standing tradition, since it contains within it the plot line found in the Sumerian Eridu Genesis story and the Babylonian Epic of Creation. Atrahasis has been found in a variety of locations, in several different versions ranging from Old Babylonian to Neo-Assyrian, and parts of a version of it were incorporated into Gilgamesh, told by Utnapishtim to the hero.30 Fragments have been found at Hattusa (there in Middle Bab- ylonian and Hittite), so Hittite scribes and educated poets were surely aware of it. The Ebla plot line in the Song of Release utilizes many of the narrative sequences found in Atrahasis: one set of beings op- pressing another with coerced labor, decisions reached in as- sembly, and interaction between a man and his god. However, in the Akkadian epic, the oppressors are gods, who decide to free another set of gods only because they can create a new set of human slaves to replace them. I have suggested that in the Song of Release, a conquering city exploits a conquered one. Al- though in both Atrahasis and the Song of Release a human has an especially close connection to a god, the relationship between god and human is portrayed differently in each poem. In Atrahasis, the eponymous hero is wholly supported by Enki; in the Song of Release, Meki does seem to have a special relationship with Teš̌š̌ub, since he presents before him the results of the decision taken by the Eblaite assembly (and probably originally reported to the assembly the god’s demands, see KBo 32.19), but he is unable to protect his city against the wrath of Teš̌š̌ub. Atrahasis opens with an untenable situation: the Anunna-gods are forcing the Igigi-gods to do corvée labor, digging irrigation ditches. The Igigi-gods attack Enlil’s house (Old Babylonian A 29 See West, East Face 489–493. 30 See Foster, Before the Muses 160. 148 THE EASTERN MEDITERRANEAN EPIC TRADITION I 39 ff.), and this triggers a debate among the Anunna-gods, who gather in assembly, in which Enlil wonders why the Igigi- gods are angry (I 99–123).31 In Atrahasis therefore the attack triggers the debate, while in the Song of Release the result of the debate seems to trigger the attack. Unlike the assembly of Ebla, the assembly of the gods decides to take effective action to re- solve the situation, deciding that the Anunna-gods must kill one of their own as a scapegoat and thus create man to relieve the Igigi-gods of their labor. I have analyzed how the Song of Release reworks particular motifs selected from the Mesopotamian narrative tradition, also found in the West Semitic literature. In the Song of Release the theme of meeting the physical needs of a superior being, which appears earliest in the Akkadian materials as the impetus for the invention of humans, is continued in the idea of a re- ciprocal relationship between gods and humans: humans can relieve the suffering of the gods by feeding and housing them. Further, it contains a message to the Hittite nobility to obey their king. However, the assembly scene of the Song of Release is more similar to the Iliad than it is to Atrahasis or any of the other Near Eastern examples. In both cases humans debate the release of captives in acrimonious terms and the unwise but powerful speaker prevails, while the other speaker is unable to carry his side even though he has a better understanding of what the god demands and the danger of opposing him. A god is angered over the captive who is not freed, and the humans debate in assembly whether the god is in fact angered because of their neglect. The humans in possession of the captives argue that they cannot do without their services, and demand that their opponents do without their own helpers instead. The Song of Release also provides an earlier literary parallel to the over- arching Iliadic narrative of the capture and return of Helen, although the rape of Helen is imbedded in a Greek set of con- cerns, of hospitality violated, gaining wives by capture, and the warrior’s honor. What are the implications of the Song of Release for those scholars who are attempting to piece together the prehistory of 31 Transl. Foster, Before the Muses 161–166. MARY R. BACHVAROVA 149 the Homeric tradition? On the one hand it illustrates how a single epic tale was transmitted across the barriers of time, space, culture, and language, from the North Syrian Hurrian sphere to the Hittite sphere.32 This presents us with a concrete example from which we can choose to extrapolate how a necessarily bilingual poet could have translated and adapted a Near Eastern epic tale, blending it with his native Greek tra- dition. We can see that “Homer” and the “singer” who was re- sponsible for the composition of the Song of Release utilized the same process to alter and adapt old themes and motifs to a new purpose. On the other hand it provides a context against which we can appreciate what is original to the Greeks and what is in- herited from the Near East in the Homeric tradition. Thirdly, the evidence of the Song of Release, when combined with the evidence from other Hurro-Hittite songs, allows us to argue that the SIR3 tradition was a key influence on the Homeric tradition within Anatolia. The Song of Release is the product of a Hurrian who belonged to the class of scribes and performers who also brought to the Hittite court in Anatolia the administrative and priestly tech- niques of North Syria. Other scribes and performers of this sort brought the Hurrian and Hittite versions of Gilgamesh; the Kumarbi Cycle (including Song of Silver, Ullikummi, and Hedammu), found only in Hittite; and the Song of Keš̌š̌i, found at Hattusa in Hurrian and Hittite (with a further Hittite version found at Ugarit). In fact, the Song of Release was found in the same build- ing as some fragments of Gilgamesh in Middle Babylonian.33 It has long been known that Gilgamesh and Kumarbi must be early examples of the tradition that lay behind the portrait of Achilles in the Iliad and Hesiod’s Theogony respectively. The Akkadian Gilgamesh borrows and reworks themes and motifs found in the earlier Sumerian literature. Thus, it inverts the oppression by corvée labor theme found in Bilgames and Akka; it 32 I discuss the evidence for adaptation of the story line to Hittite interests in JAOS (forthcoming). 33 See G. Wilhelm, “Neue akkadische Gilgameš -Fragmente aus Hattu š a,” ZA 78 (1988) 99–121, at 100–101. 150 THE EASTERN MEDITERRANEAN EPIC TRADITION includes the flood story from a version of Atrahasis;34 and it uses the creation of man sequence to introduce Enkidu.35 In the same way, the Hurro-Hittite Kumarbi reuses elements of the succession of gods theme, but also adds new elements, ones found in Hesiod’s Theogony and in Phoenician works which the Greeks may have known. We do not have enough of the Hittite version of Gilgamesh to see to what degree it was altered, al- though what we do have shows that it was tailored especially for an audience who resided to the west of Mesopotamia; some parts are abbreviated, while Gilgamesh’s trip to the Cedar Mountains was developed at further length. (These mountains, after the Old Babylonian period, were thought to lie in the direction of Anatolia rather than to the east.)36 The Song of Release reuses many elements of Atrahasis but moves most of them to the human sphere, just as the Iliad combines and adapts elements found in the Near Eastern tradition to speak to the concerns of an emergent polis structure, in which the per- sonal glory of warriors was still of paramount importance, and reciprocal relations between men were still conceived of in terms of exchange of women, gifts, and compensation. It is remarkable that three separate SIR3’s found at Hattusa, the Song of Release, Gilgamesh, and Kumarbi, should have such close ties to hexametric poetry in the Ionic dialect. This, along with other evidence, argues for Anatolia as an important location in which Greek-speakers were made aware of Near Eastern epic and incorporated elements of it into their native Indo-European narrative poetry glorifying gods and men. Ana- tolia is the site of the action of the Iliad and the area in which the Ionicized Homeric dialect was developed. A New Hittite treaty (approx. 1200 B.C.) names as the ruler of “Wilusa,” which many have equated with “(W)ilios,” a man named “Alaksandu,” among whose gods appears a certain “Appali- unas,” although there is a frustrating break at the beginning of 34 See J. H. Tigay, The Evolution of the Gilgamesh Epic (Philadelphia 1982) 214–240. 35 Tigay, Evolution 192–197. 36 Tigay, Evolution 111–118. MARY R. BACHVAROVA 151 his name which still leaves slight uncertainty.37 If we take this Anatolian god to be Apollo, the Iliad is accurate when it names him as the patron god of Ilios.38 Furthermore, Anatolia is the only location in which actual transfer across languages of a motif we consider to be Homeric is found, as attested by a single line of a Luwian song about “steep Wilusa” found in a Hittite festival text. Just as Hurrian-speakers brought the SIR3 genre to Hattusa, priests and cult performers certainly carried back and forth songs from Hattusa to Troy, where Greek- speakers mingled with native Anatolians.39 We also know that 37 The treaty is translated by G. Beckman, Hittite Diplomatic Texts2 (Atlanta 1999) 87–93. Note that the handcopy of the Alaksandu treaty is incorrect, and therefore the editions and translations of the text are incorrect. My examination of a photograph of the tablet, kindly provided by H. A. Hoffner, Jr., shows that the remains of the broken sign before ap-pa-li-u-na- aš̌ (Keilschrifturkunden aus Boghazköi [KUB] 21.1 iv 27) are consistent with the “god” determinative. The topic of contact between Greek-speakers and Anatolians in the second millennium has been fraught with controversy ever since E. Forrer first suggested it, and F. Sommer denied it: see O. Szemerényi, “Hounded out of Academe…: The Sad Fate of a Genius,” in F. Imparati (ed.), Studi di storia e di filologia anatolica dedicati a Giovanni Pugliese Carratelli (Florence 1988) 257–295. Among the points of contention has been whether Wilusa should be located in the Troad, a question difficult to resolve given the paucity of knowledge concerning the political geography of western Anatolia. The plausible deciphering of a Hieroglyphic Luwian monument marking the southern edge of Mira has greatly improved our knowledge and supports placing Wilusa in the vicinity of Troy: J. D. Hawkins, “Tarkasnawa King of Mira: Tarkondemos, Boğazköy Sealings and Karabel,” AnatSt 48 (1998) 1–32. 38 R. S. P. Beekes, “The Origin of Apollo,” JANER 3 (2003) 1–21, has recently presented a strong argument against W. Burkert’s claim (“Apellai und Apollon,” RhM 118 [1975] 1–21) that Apollo’s name has a Greek etymology. Also see Schuol, Grenzüberschreitungen 350, on the early attesta- tions of Apollo in Cyprian and Mycenaean. 39 C. Watkins, “The Language of the Trojans,” Selected Writings (Inns- bruck 1998) 700–717, in 1986 was the first to note the significance of the single line of Luwian song mentioning Wilusa. The multilingual context of Mycenaean Age Troy has already been well-discussed by Watkins, “Lan- guage of the Trojans,” and “Homer and Hittite Revisited,” in P. Knox and C. Foss (eds.), Style and Tradition: Studies in Honor of Wendell Clausen (Stuttgart 1998) 201–211, and F. Starke, “Troia im Kontext des historisch-politischen 152 THE EASTERN MEDITERRANEAN EPIC TRADITION gods were transported from Lazpa (Lesbos) and Ahhiyawa (Achaia) to the Hittite capital. This would have provided op- portunities for the transmission of songs honoring gods to cross linguistic barriers.40 We have then at least two periods in which transfer could have taken place, one in the second millennium and one in the first, when the Homeric dialect reached its final form. Besides these attested points of contact, there were op- portunities for many more in this part of the eastern Mediter- ranean, at Miletus, Cyprus, Sardis, and Ugarit, for example, where Greek-speakers in the Mycenaean period would have had an opportunity to hear and be inspired by the Hurro- Hittite SIR3 tradition. ___ und sprachlichen Umfeldes Kleinasiens im 2. Jahrtausend,” Studia Troica 7 (1997) 447–487. It should be noted however that there is no strong evidence that Luwian was the Anatolian language spoken at Troy: H. C. Melchert, “Prehistory,” in H. C. Melchert (ed.), The Luwians (Leiden 2003) 8–26, at 11–12 (with earlier references), and M. Hutter, “Aspects of Luwian Re- ligion,” in Luwians 211–280, at 265–266. (Hutter remains skeptical of the significance of the so-called Ahhiyawa material, see 264–269.) J. Puhvel has discussed the linguistic evidence for such contact between Anatolian and Greek-speaking poets in several articles: “Homeric Questions and Hittite Answers,” AJP 104 (1983) 217–227, “An Anatolian Turn of Phrase in the Iliad,” AJP 109 (1988) 591–593, Homer and Hittite (Innsbruck 1991), “A Hittite Calque in the Iliad,” Historische Sprachforschung 106 (1993) 36–38. Schuol, Grenzüberschreitungen 345–353, again reviews the evidence of the contact between Mycenaean Greek-speakers and Anatolians, arguing that evidence from the Song of Release has implications for the transition from oral to written of Homeric epic. P. Högemann, “Zum Iliasdichter—Ein anatoli- scher Standpunkt,” Studia Troica 10 (2000) 183–198, argues that Homer composed for an Anatolian audience in the first millennium. See on the other hand T. R. Bryce, “Anatolian Scribes in Mycenaean Greece,” Historia 48 (1999) 257–264, speculating that Anatolian scribes brought the SIR3 tradition to Crete and the Mycenaeans. 40 On the transfer of gods see S. P. Morris, “Potnia Aswiya: Anatolian Contribution to Greek Religion,” in R. Laffineur and R. Hägg (eds.), Potnia: Deities and Religion in the Aegean Bronze Age (Liège 2001) 423–434. As she notes (following Watkins, Writings 700–717), a Hittite oracle text (KUB 5.6 ii 57', 60') asks how king Mursili II should welcome the gods of Lazpa and Ahhiyawa. She suggests further that the mention of a Potiniya Asiwiya in a Mycenaean text from Pylos (Pylos Fr 1206) indicates that a goddess from Assuwa, in Western Anatolia, made the journey from Anatolia to the Greek mainland. MARY R. BACHVAROVA 153 Although Greek epic could have had the opportunity to draw on Anatolian versions of Mediterranean epic, there is no reason to assume that Homer or one of his ancestors directly imitated the Song of Release, any more than we should assume that they directly imitated a version of Gilgamesh preserved for us. All these songs are drawing on a wider tradition of which only a few examples are preserved, whether from the Myc- enaean period or from the Archaic period. However, it is safe to surmise that Homeric poets, at some point in history, were in contact with an offshoot of the Near Eastern epic tradition which was particularly close to the SIR3 tradition preserved for us at Hattusa.41 February, 2005 Dept. of Classical Studies Willamette University 900 State St. Salem, OR 97301 mbachvar@willamette.edu 41 This paper is an updated and streamlined version of a section of my dissertation From Hittite to Homer. The material has been presented at meet- ings of the American Oriental Society (2000), the American Philological Association (2001), and the Classical Association (2001). I must thank H. A. Hoffner, Jr., and C. Watkins for all their help, although my views should not be imputed to them.